4

According to the reputation audit, for the latter half of yesterday my counts were:

rep cap was reached on 50 days
rep cap was exceeded on 50 days

I didn't get Epic right away, and I thought maybe I'd have to wait until the end of the day to get it. To start today, my counts are:

rep cap was reached on 50 days
rep cap was exceeded on 51 days

I still don't have an Epic. Is this a bug, or have I still not qualified somehow?

2
  • For what it's worth, the old greasemonkey epic/legendary badge script from Shog9 that runs off the reputation chart data says that you have 4 more days to go. There are probably discrepancies between the rep audit and the actual mortarboard count in the system.
    – Aarobot
    Commented May 13, 2010 at 2:53
  • 1
    @Aarobot: one would think that /reputation is the most up-to-date, authoritative, reliable voice on these issues. But perhaps one would think wrong. Commented May 13, 2010 at 4:37

1 Answer 1

3

Rarely awarded badges aren't calculated very often for performance reasons. Wait 2 full days, and if you don't have it by then let us know.

4
  • 1
    You can perhaps have a special "maybe today" group, e.g. those that have 49, and check members of that group every 4 hours or something, see if they finally reach 50. It's true that there may be no bound on the size of this group, but I'm guessing most people at 49 wouldn't want to stick around in the "maybe" zone for too long =) Commented May 13, 2010 at 6:07
  • @poly: Then instead of doing one computation (#DAYS ≥ 50), the system needs to do two (#DAYS ≥ 49, #DAYS ≥ 50 && in Maybe zone).
    – kennytm
    Commented May 13, 2010 at 7:50
  • 2
    @KennyTM: if performance is that much of an issue, you can put people in even more buckets. Let's say there's the [0..25) bucket A, the [25..40) bucket B, and the [40..50) bucket C. Check bucket C every day, see who qualifies for {Epic}. Check bucket B every 10 days, see who migrates to bucket C. Check bucket A every 15 days, see who migrates to bucket B. Bucket A is largest (presumably), but also checked the most infrequently. You can play around with the parameters to see what works best for actual distribution. Commented May 13, 2010 at 10:26
  • just got it. Feel free to [status-completed], or even delete if this is too localized. Commented May 13, 2010 at 12:30

Not the answer you're looking for? Browse other questions tagged .